Adolescent sexual offenders present serious clinical problems that have led to expensive (and untested) interventions from mental health and juvenile justice authoritics. Based on the extant literature, we argue that the predominant (i.e., restrictive) interventions for adolescent sexual offenders are not well matched with the developmental needs of these youths; and, consequently, these interventions are not likely to be more effective than community-based alternatives. Moreover, we suggest that effective services will (a) have the capacity to address multiple risk factors; (b) be individualized; (c) overcome barriers to service access; (d) include strong quality assurance mechanisms; (e) focus on building the capacities of the environmental systems, especially the family, surrounding the youth; and (f) promote community safety. To test these contentions, this application proposes an effectiveness study comparing adolescent sex offenders treated with multisystemic therapy (MST; Henggeler, Schoenwald, Borduin, Rowland, & Cunningham, 1998) versus adolescents treated with usual juvenile justice and mental health services, In a randomized design, 160 juvenile sex offenders will be assigned to MST (N = 80) vs. Usual Services (US; N = 80) treatment conditions. A multimethod, multirespondent assessment battery will be used to examine key outcomes pertaining to clinical (e.g., criminal behavior, symptomatology ) and service (e.g., service utilization, cost effectiveness) outcomes. Assessments will be conducted at 6-month intervals through a 2-year post referral follow-up, with monthly tracking of service utilization and treatment adherence. The specific aims of the study are:Aim 1. To evaluate treatment effects on youth criminal activity, mental health functioning, and alcohol and drug use," as well as on family relations, peer relations, and school attendance.Aim 2. To track adolescent juvenile justice involvement, mental health and substance abuse service utilization, and out-of-home placements and their associated costs across the treatment conditions for 24 months post recruitment.Aim 3. Assuming favorable MST treatment effects, to examine the possible moderators (e.g., race, sexual offender subtype) and mediators (e.g., improved family relations) of such effects.